Lab-scale preparation of a novel carbocyclic chemokine receptor antagonist

Article abstract of DOI:10.1021/op500265z

The preparation of a novel chemokine receptor type 2 (CCR-2) antagonist is described on a 135 g scale. The synthesis of an all-carbon bicyclic core was accomplished using a radical cyclization strategy using chiral precursors, wherein elaboration led to N-Boc carboxylic acid in good yield. After amidation using a traditional coupling reaction, a reductive amination using enantiomerically enriched 3-methoxy-4-pyranone led to the final compound. Although several steps of the syntheses involved reagents that would not be preferred in process and chromatography was used to provide the free-base diastereomer of the final succinate salt, the overall route went through stable intermediates that could be used for future scale-up. This lab-scale synthesis struck a balance between a quick scale-up and a more thorough process review of all possible methods and routes.

Full text of DOI:10.1021/op500265z

Article
pubs.acs.org/OPRD
Lab-Scale Preparation of a Novel Carbocyclic Chemokine Receptor
Antagonist
Christopher A. Teleha,*,† Shawn Branum,† Yongzheng Zhang,† Michael E. Reuman,† Luc Van Der Steen,‡
Marc Verbeek,‡ Nagy Fawzy,§ Gregory C. Leo,∥ Fu-An Kang,⊥ Chaozhong Cai,⊥ Michael Kolpak,∥
Derek A. Beauchamp,† Mark J. Wall,⊥ Ronald K. Russell,† Zhihua Sui,⊥ and Hilde Vanbaelen‡
§
∥
⊥
†High Output Synthesis, Pharmaceutical Sciences, Analytical Research, and Cardiovascular and Metabolic Disease Research,
Janssen Pharmaceutical Research and Development, Welsh and McKean Roads, P.O. Box 776, Spring House, Pennsylvania 19477,
United States
‡Preparative Separation Techniques, API Small, Janssen Research and Development, 30 Turnhoutseweg, Beerse, B-2340, Belgium
S
* Supporting Information
ABSTRACT: The preparation of a novel chemokine receptor type 2 (CCR-2) antagonist is described on a 135 g scale. The
synthesis of an all-carbon bicyclic core was accomplished using a radical cyclization strategy using chiral precursors, wherein
elaboration led to N-Boc carboxylic acid in good yield. After amidation using a traditional coupling reaction, a reductive
amination using enantiomerically enriched 3-methoxy-4-pyranone led to the final compound. Although several steps of the
syntheses involved reagents that would not be preferred in process and chromatography was used to provide the free-base
diastereomer of the final succinate salt, the overall route went through stable intermediates that could be used for future scale-up.
This lab-scale synthesis struck a balance between a quick scale-up and a more thorough process review of all possible methods
and routes.
INTRODUCTION
■
Antagonists of the CC-chemokine receptor 2 (CCR-2) have
been vigorously pursued by a number of pharmaceutical
companies as a target for drug discovery, in that compounds
could have the potential for use in the acute and chronic
conditions of inflammatory and autoimmune diseases asso-
ciated with infiltration of monocytes, macrophages, lympho-
cytes, dendritic cells, NK cells, eosinophils, basophils, natural
killer (NK cells), and memory T-cells. A compound of interest
that was discovered in the Janssen laboratories that met the
initial criteria set out during the in vitro screening phase of drug
discovery was bicyclic 1.1,2 The Discovery Chemistry team
charged us with up-scaling the preparation of 1 to 135 g for use
in a 14-day tox/tolerability study. Although we realized that we
may have to incorporate some process-unfriendly techniques,
we had to do a meld of discovery/process that matched the best
of pure speed of preparation with work-arounds to allow us to
Figure 1. Retrosynthesis of CCR2 antagonist compound 1.
start thinking of a future process. For the scale-up of 1, we took
advantage of the process innovations which we uncovered
during the course of our investigations and are reported here.
The Discovery Chemistry retrosynthesis of 1 in Figure 1 was
product 2 with ∼10% of two diastereomers, in the hope that
A time factor that had to be considered was chiral
chromatography of the reductive amination product 2. Several
quick attempts were made to form salts of the crude reduction
reasonable for discovery scale-up and entailed: (1) amide
formation between Boc acid 4 with fused piperidine 5; (2)
deprotection of the NHBoc protecting group; (3) reductive
amination of the newly revealed amine with the chiral ketone 3;
(4) chiral separation of the reductive amination products to
provide diastereomerically pure 1 as the free base; and (5) salt
formation with succinic acid to give final compound 1. This
general synthesis strategy of these final steps to arrive at 1 could
not be altered under the time constraints of the project.
crystallization could render the product free of the diastereo-
meric contaminants (described later). However, there were no
hits or enrichment of pure diastereomer 2 in this investigation,
so chiral chromatography of the reductive amination product
was accepted as the only sure-fire method to obtain
diastereomerically pure product. Several preparations of the
Received: July 17, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/op500265z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 1. Discovery Chemistry route to all-carbon core acid 4
Figure 2. Background of bicyclic ethers and alkanes and proposed disconnection for 10.
Scheme 2. Attempted radical cyclization for the formation of 16
a
b
c
I(CH2)3Cl, LiHMDS, THF, −20 to 0 °C (66%). Nal, acetone, 60 °C (90%). AIBN/(TMS)3SiH or Bu3Sn/benzene, 80 0 °C.
enantiomer of 3 have been described,3 and so access to ketone
3 was expected with some slight variations.4 The chiral purity of
3 that we had was ∼86−95% ee, thus leading to the earlier
described diastereomers of 2. This confirmed the decision to
use chiral chromatography to arrive at diastereomerically pure
2.
The acid 4 was previously described in both chiral and
racemic form.1 The absolute stereochemistry of 4 was based on
an X-ray crystal structure performed on the final compound 1
and confirmed the positional relationships of the groups on a
[3.3.0]-bicyclic carbon core. Although the chiral route (Scheme
1) was justifiable to obtain acid 4 and the precursor ester 10 in
chiral form on a discovery scale, the reductive cleavage of the
exo-olefin was cumbersome and not amenable to scale-up. This
3 + 2 cycloaddition strategy, as well other work in the area,5
demonstrated that the −NHBoc group was effective to direct
chemistry to the anti- face of the cyclopentene ring, and so we
looked into other prominent ring-closing strategies for the
formation of bicyclic ring systems.
the pioneering work of Beckwith and Roberts,6 utilizing alkyl
bromide 12 and a trialkylstannane/free-radical initiator method
(Figure 2). In addition, this general strategy has been used to
synthesize closely related bicyclic systems.7 It is easy to see the
radical cyclization method applied to iodide precursor 13,
although use of trialkylstannane reagent would not be
acceptable for any scale-up. Hence, the synthesis of ester 10
represented a unique challenge with the nitrogen side chain
protecting group.
Chiral nitrogen protected cyclopentenes have been featured
as intermediates in the syntheses of nucleosides as inhibitors of
γ-aminobutyric acid aminotransferase and as antagonists of
chemokine receptor antagonists.8,9 In particular, pyrrole
protected 143c has been used as an intermediate in the
preparation of MK-0812,10 because of its ability to provide
excellent diastereoselectivity in the alkylation with electrophiles.
We confirmed the high diastereoselectivity seen in the
alkylation of 14 upon treatment with 1.6 equiv of LHMDS,
followed by 3-chloro-1-iodopropane, which, after the Finkel-
stein reaction, provided iodide 15 as the necessary precursor for
radical cyclization (Scheme 2). The reaction sequence gave
Bicyclic alkanes, such as the [4.3.0] ring system exemplified
by 11, have been prepared by free-radical cyclization based on
B
dx.doi.org/10.1021/op500265z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 3. Prepartion of acid 4
a
b
c
d
I(CH2)3Cl, LiHMDS, THF, −70 to 0 °C (25−65%, see discussion). Nal, acetone, 60 °C (96%). TMS3SiH AIBN, PhMe, 80 °C (63%). 10N
NaOH, MeOH, H2O, 60 °C (93%).
Scheme 4. Steps of final compound 1
predominantly one isomer which was used to investigate the
downstream chemistry.
Upon subjecting 15 to radical cyclization conditions (AIBN/
(TMS)3SiH or Bu3SnH/benzene/80 °C), we failed to isolate
any 16, only detecting deiodinated starting material by LCMS,
probably due to radical quenching by the electron-rich pyrrole
group.11 The N-phthalimide protecting group as a replacement
for the 2,5-dimethylpyrrole group was briefly investigated but
was unstable under the alkylation conditions. Finally we
changed to the ester 6,8 and this allowed us to complete the
synthesis. As shown in Scheme 3, N-Boc ester 6 was a
competent alkylation nucleophile, whereupon treatment with
C
dx.doi.org/10.1021/op500265z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
2.2 equiv of LHMDS gave the expected dianion that, upon
reaction with 3-chloro-1-iodopropane, provided alkylation
products 17 and 18, favoring the desired isomer in a ratio of
6:1. The same β-facial selectivity was seen for this reaction, in
concert with that observed for 2,5-dimethylpyrrole protected
14, although the diastereoselectivity was lower for the Boc
group.
The reaction worked better using an inverse addition,
whereby the preformed dianion was added to a chilled solution
of the electrophile. Yields were better with reverse addition,
even though this was more difficult operationally. Two
additional parameters were the key to reproducible yields:
(1) it was important to maintain an inert atmosphere over both
the anion and electrophile flasks during the course of the
reaction, and this was effectively accomplished with high flow
nitrogen sweeps; (2) a gradual temperature increase from −70
to 0 °C after cannulation of the anion into the electrophile was
required for good results.12 When these two precautions were
not taken, the yield dropped from 65% to 25−33%.
Side products for the early low-yielding reactions were not
investigated once the key parameters were identified. The
alkylation reaction also produced some chain-elimination
product 19; this was easy to remove relative to the close-
eluting 17 (less polar) and 18 (more polar) products on silica
gel. In the mass balance analysis, when alkylation was
incomplete, the starting material with the double bond moving
into conjugation with the ester was detected but never to such
an amount that separation was an issue. Poor results were seen
when 1,3-dibromopropane was used, favoring 19 as the major
product. Alkylation products 17 and 18 were oils on initial
isolation, although higher purity batches of 17 did slowly
crystallize to a low-melting solid on standing. No follow-up
work was done to take advantage of this property. Although
useful for this lab campaign to obtain the required materials,
this alkylation sequence will require alterations if considered for
further scaling.
The Finkelstein reaction (NaI/acetone/65 °C) of chloride
17 provided iodide 13 in high yield. Gratifyingly, the reaction of
13 under radical cyclization using tin-free radical precursor
tris(TMS)silane gave [3.3.0] alkane 10 in 63% yield. The
cyclization proceeded with the expected diastereoselectivity,
providing the cis-fused bicyclic ring, owing to the short chain
length. There may have been a small amount of trans-fused ring
product from the reaction but was not investigated to
determine if it came from trans-cyclization or if it came as a
result of contamination of 18 in the chromatographed 17
product.
solvent to DMAc, which allowed for isolation of amide 21 by
filtration from an aqueous solution of 5% sodium bicarbonate
that was used to quench and workup the reaction. A 75% yield
was obtained for the filtered coupling product, which was
considered acceptable taking into account that the revised
procedure saved time by eliminating purification by column
chromatography.
Deprotection of the Boc group in amide 21 was originally
performed using mixtures of TFA in DCM. Aside from the fact
that the TFA salt of 23 was an oil after concentration of the
reaction mixture, we observed some variation in the rate of the
reaction depending on the charge of TFA. Deprotections using
DCM mixtures of 21 with 6−20 equiv of TFA were only 80%
complete at rt overnight. Heating to 50 °C while concentrating
the reaction solution as part of the workup led the reaction to
completion. As such, we changed the deprotection conditions
to 1.25 M HCl in MeOH which rendered the bis HCl salt 22 as
a solid. The reaction was heated from 40 to 60 °C, at which
temperature the reaction did have a vigorous off-gassing. At lab
scale (220 g), nonrestrictive venting was used on the outflow of
the inert gas sweep to prevent pressure build-up. After 40 min
hold, the reaction had a clean HPLC profile for completion,
and the reaction was evaporated.
Choice of the bis-HCl salt product 22 was fortuitous in that
it was easily rendered as a filterable solid upon trituration with
i-PrOH at 40 °C. Addition of heptane facilitated higher
recoveries of the product in the ensuing filtration. The HCl salt
22 was isolated as a yellow solid in 85% yield. The color of the
solid did seem to have some variation depending on the color
of the starting 21 used for the reaction. An attempt was made
to use 5−6 M HCl in i-PrOH for the deprotection; the reaction
again required heating and was efficient in the conversion to
product, but the solid form that was initially obtained (without
the heptane treatment) was not as filterable as the solid
obtained by the HCl/MeOH method.
The reductive amination reaction between newly revealed
amine and the chiral ketone 3 was the most challenging step of
this scale-up. We were specifically worried about several aspects
of this reaction: (a) the reaction was performed in DCM, which
could have the possibility to react in some cases with amines;14
(b) chiral ketone 3 was prone to racemization in the presence
of base; (c) the reductive amination of amines with ketones are
not unprecedented but generally less prevalent than amine/
aldehyde combinations;15 (d) the reductive amination was
found to be cis-selective (with respect to the α-OMe group)
but not exclusive, meaning that there was some detectable
amounts of the trans-reductive amination product formed; and
finally, (e) the chirality of the ketone 3 was not 100% ee, thus
leading to the above-described diastereomers.
A through-process utilizing a mixture of alkylation isomers
17 and 18 directly into the radical cyclization was considered
but was ruled out since, under the time constraints of the
project, it was too risky to rely on one chromatography column
to effectively separate isomers at the methyl ester stage. With a
single isomer ester 10 in hand, hydrolysis of the ester gave acid
4 in 93% yield. The sequence was viable towards accomplishing
our goal, with the scalability of the reactions highlighted by a
radical cyclization reaction conducted on a 385 g scale.
With quantities of acid 4 in hand, the final steps were
identical to that reported by the Discovery Chemistry team.
Amide formation between acid 4 and fused piperidine 513
utilized a standard reagent combination of EDC/HOBt/Et3N
in THF and required concentration before workup. We felt that
the reagents were satisfactory for scale-up, incorporating only a
minor change in bases (Scheme 4). We did change the reaction
Addressing these concerns in this reaction, we reasoned that
changing the DCM solvent to 1,2-dichloroethane (DCE) was
an acceptable replacement that would not alter the overall
reaction. The discovery team had performed the reductive
amination using the TFA salt of amine 23 and excess tertiary
amine which caused an increase in the unwanted diastereomers
from ketone racemization. To eliminate this complication, we
chose to free-base the HCl salt 22 as a separate operation;
ideally the free-basing would be done using DCE in the
extraction step, so a through-process could be used. The
process worked well except that we encountered a heavy rag
interface between the DCE and the aqueous NaOH used for
the neutralization. This required a filtration of the DCE layer
through Celite to achieve a clear organic solution. After drying
D
dx.doi.org/10.1021/op500265z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
the DCE layer with drying agent, Karl Fischer titration
indicated the organic layer had ∼2% water which was not
controlled to any strict level. The reductive amination was
carried out using the DCE solution of free-base 23, along with
glacial AcOH (2 equiv), sodium triacetoxyborohydride (STAB)
(1.4 equiv) followed by charging a DCE solution of 4 (1.4
equiv) at 0 °C. After 1 h, some 23 was still found to be present,
so an additional 5% charge of 4 was made before the reaction
was allowed to warm to rt overnight. The presence of the trans-
product 24 was detected at 2.5% in the first hour of the reaction
and had increased to 4.6% at the end of the reaction (24 h).
The reaction was stopped with <0.5% 23 left unreacted. Also
detected in the crude reaction product was ∼5% of 25, which
was present as a consequence of the ketone enantiomer present
in the batches of 4 used for the scale-up. We did not have
access to enantiopure 4, so the presence of 25 was
understandable but also unavoidable. The presence of
diastereomer 26 was theoretically possible based on the
reactants but was not quantitated in the crude product mixture.
As discussed earlier, chiral chromatography rendered the pure
diastereomer 3 in 97.6% de and in 75% yield as an oil that
retained about 3% w/w EtOH.
Polymorph screening of the succinate salt 1 revealed that
acetone was a good solvent to provide a crystalline solid (as
judged by XRPD) that was required for testing in the tox/
tolerability studies. The salt formation procedure was
straightforward: a heated solution of 3 with an eqimolar
equivalent of succinic acid which resulted in a clear solution,
which on seeding at 50 °C was allowed to cool to rt. After
chilling to 0 °C, filtration, washing and drying provided the
product. The succinate salt 1 directly obtained using chromato-
graphed amine was ivory-colored, despite some color being
retained in the mother liquors and was not brilliant-white as
seen in earlier lots, so a decolorization procedure was required.
The salt 1 was dissolved in MeOH, treated with 20% w/w
Darco-60 charcoal, and warmed to 50 °C. Filtration and
concentration gave a foam that was again recrystallized from
acetone (6 vol). Seeding was not required, and upon cooling/
filtration/washing and vacuum oven drying, there was obtained
1 in 81% yield in the chiral purity of 99.5% de and 99.1%
HPLC purity. The API met the team’s requirements for use in
the tox/tolerability studies. Results of these tests will be
reported elsewhere.
of other complex CCR-2 antagonists can be found in our
companion paper.2
EXPERIMENTAL SECTION
■
(1R, 4S)-Methyl-1-(3-chloropropyl)-4-(2,5-dimethyl-
1H-pyrrol-1-yl)cyclopent-2-enecarboxylate (15). To a
solution of LHMDS (1 M in THF, 36.5 mL, 36.5 mmol, 1.6
equiv) cooled to −20 °C was added a solution of 143c (5 g,
22.8 mmol) in THF (23 mL) over 20 min. The anion was
stirred for 20 min at −20 °C, cooled to −70 °C (dry ice/
acetone), and 1-chloro-3-iodopropane (3.4 mL, 31.6 mmol, 1.4
equiv) was added quickly and the resulting solution stirred for 1
h at −70 °C. The bath was replaced by an ice bath, and the
temperature was allowed to rise to 0 °C and stirred for 3 h at
this temperature. The reaction was quenched by addition of
aqueous NH4Cl (6%, 25 mL) and isopropyl acetate (iPAc, 50
mL). The organic layer was separated and washed with aqueous
NH4Cl (6%, 25 mL) and brine (2 × 25 mL). The organic layer
was dried (Na2SO4), filtered, and evaporated and gave (1R,4S)-
methyl-1-(3-chloropropyl)-4-(2,5-dimethyl-1H-pyrrol-1-yl)-
cyclopent-2-enecarboxylate (4.42 g) as a brown oil in 66%
1
yield. H NMR (400 MHz, CDCl3) 6.03−6.05 (m, 2 H), 5.74
(s, 2 H), 5.30 (s, 1 H), 3.68−3.75 (m, 3 H), 3.54 (t, J = 6.2 Hz,
2 H), 2.30−2.49 (m, 2 H), 2.20 (s, 6 H), 1.67−1.91 (m, 2 H);
13C NMR (100 MHz, CDCl3) 175.22, 134.56, 133.51, 128.17,
106.35, 60.07, 58.52, 52.21, 44.82, 40.34, 35.83, 28.80, 13.83;
LCMS (m/z): 296 (M + H, 100); exact mass calc for
+
C16H23ClNO2 : 296.1412, found 296.1410.
To part of the above isolated chloride (1.15 g, 3.89 mmol) in
acetone (14 mL) was added NaI (2.90, 19.4 mmol, 6 equiv),
and the mixture was heated to 60 °C for 5.5 h. The reaction
was cooled to rt and diluted with water (20 mL) and EtOAc
(25 mL). The layers were separated, the aqueous layer
extracted with EtOAc (20 mL), and the combined organics
washed with brine (25 mL) and dried (Na2SO4). After
filtration, the volatiles were removed on a rotary evaporator
(29 in Hg, 40 °C), and the product 15 (1.36 g) was isolated as
a red oil in 90% yield. 1H NMR (400 MHz, CDCl3) 5.97 (s, 2
H), 5.74 (s, 2 H), 5.30 (t, J = 8.3 Hz, 1 H), 3.69−3.77 (m, 3
H), 3.13−3.21 (m, 2 H), 2.31−2.51 (m, 2 H), 2.20 (s, 6 H),
1.76−1.87 (m, 2 H);); 13C NMR (100 MHz, CDCl3) 175.15,
134.56, 133.52, 128.16, 106.33, 60.01, 58.38, 52.22, 40.30,
39.24, 29.45, 13.82, 5.95; LCMS (m/z): 388 (M + H, 100);
+
exact mass calc for C16H23INO2 : 388.0768, found 388.0621.
Alkylation of Ester 6 with 3-Chloro-1-iodopropane.
LHMDS in THF (1M, 1.24 L, 1.24 mol) was quickly measured
into a graduated cylinder under vigorous stream of N2,
transferred to a reaction flask, and cooled to −73 °C. A
solution of 68 (136 g, 0.564 mol) in THF (544 mL) was added
at < − 67 °C. The resulting orange anion solution was stirred
for 1.5 h at −70 °C. Another flask was charged with 3-chloro-1-
iodopropane (85 mL, 0.789 mol) in THF (136 mL) and cooled
to −69 °C. The orange anion solution was transferred over 30
min while keeping the temperature about −55 °C. The reaction
was warmed to −40 °C and held for 30 min, −20 °C and held
for 30 min, and then 0 °C for 30 min. Progress of the reaction
was evaluated by TLC and NMR (quench aliquot): (20%
EtOAc/heptanes, Rf 17 = 0.1; Rf 18 just below 17; Rf 19 just
above 18; Rf 6 (conjugated) just above 19; Ceric Ammonium
Molybdate) [If the reaction is dark green or dark brown in
color at this point, typically this was the result of an inadequate
nitrogen sweep]. The reaction was carefully poured into a 4 L
beaker which contained ice (750 g) and aqueous HCl (2 M, 1
CONCLUSION
■
We have described the scale-up of CCR-2 antagonist 1 as its
succinate salt in 135 g quantity that met the purity and
polymorph specifications and achieved the goal of making the
required amount within an aggressive timeline. The process for
the formation of 1 involved a new method relative to the initial
discovery chemistry route and focused on the preparation of all-
carbon NHBoc acid 4 in 36% overall yield from starting 6. The
process involved diastereoselective enolate alkylation that led to
an iodide precursor 13 that underwent a radical cyclization
using tin-free conditions to intersect the previously known 10.
The final steps involved a sequence identical to the discovery
route, coupling of acid 4 with amine 5, Boc deprotection of 21,
reductive amination with 3, chromatographic enrichment for
the desired diastereomer 2, and succinate salt formation
completed the synthesis of final compound 1 in 8 steps overall
and 14% yield from 6. Additional work in the exploration of
alternate cyclization strategies directed toward the preparation
E
dx.doi.org/10.1021/op500265z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
L). The layers were separated in a separatory flask, and the
aqueous was extracted with toluene (2 × 500 mL). The organic
layers were combined, washed with H2O (1 L and 500 mL) and
brine (1 L and 500 mL), dried over Na2SO4, filtered, and
evaporated (rotary evaporator, 29 in Hg, 40 °C) to afford 169 g
of crude red oil. Purification by flash chromatography (2.5 kg)
using the following conditions: a prewetted column with
heptane (4 L) was loaded with crude product in CH2Cl2 (100
mL), eluted with heptane (4 L), 10% EtOAc/heptane (16 L),
and 15% EtOAc/heptane (16 L). There was provided 17
(116.7 g, 65%) as a slowly crystallizing solid on standing; 18
was not isolated in this run but typically isolated in a mass
recovery of 25%. The NMR ratios of 17/18 were seen in the
range of 3.8:1 to 16:1 depending on the scale and variability of
the run. 19 was only isolated for identification purposes and not
fully characterized.
mol, 1.28 equiv), and toluene (850 mL) and stirred until
homogeneous. This homogeneous solution was added
dropwise to the reaction flask over 1 h at >82 °C. The
reaction was stirred for 2 h. (The progress of the reaction was
monitored by TLC (20% EtOAc/heptanes, Rf dilane debris =
0.6; Rf 10 = 0.4; Rf cyclization product from iodo-18 = 0.33;
KMnO4 stain). The reaction was cooled to rt, washed with H2O
(2 × 4 L), brine (2 L), dried over Na2SO4, filtered, and
concentrated to afford crude product (681 g) which contained
toluene and TMSI. The crude material was split in two equal
portions (loaded in mix of DCM and heptane) and each
purified by flash chromatography (5 kg, prewetted with
heptane, 8 L) using a isocratic solvent system of 10%
EtOAc/heptane and provided 10 (166.7 g) as a clear oil in
63% yield. 1H NMR (400 MHz, CDCl3) 4.88 (br. s., 1 H), 4.08
(d, J = 7.1 Hz, 1 H), 3.69 (s, 3 H), 2.69−2.84 (m, 1 H), 2.13−
2.24 (m, 1 H), 2.08 (dd, J = 7.0, 13.1 Hz, 1 H), 1.71−1.95 (m,
4 H), 1.48−1.71 (m, 3 H), 1.33−1.47 (m, 10 H); 13C NMR
(100 MHz, CDCl3) 179.3, 155.5, 79.1, 58.3, 52.2, 52.1, 47.5,
(1R,4S)-Methyl-4-(tert-butoxycarbonylamino)-1-(3-
chloropropyl)cyclopent-2-ene-carboxylate (17). mp
1
52.8−61.3 °C, H NMR (400 MHz, DMSO-d6) 6.84 (d, J =
25
7.6 Hz, 1 H), 5.73 (dd, J = 2.1, 5.5 Hz, 1 H), 5.56−5.65 (m, 1
H), 4.37−4.50 (m, 1 H), 3.45−3.56 (m, 5 H), 2.07 (dd, J = 7.9,
13.6 Hz, 1 H), 1.88 (d, J = 6.8 Hz, 1 H), 1.43−1.65 (m, 1 H),
1.28 (s, 9 H); 13C NMR (100 MHz, DMSO-d6) 175.03, 155.00,
134.47, 133.89, 77.68, 57.72, 51.85, 45.24, 35.27, 28.21, 28.15;
43.6, 40.3, 38.7, 34.0, 28.4, 26.2; [α]D +17.13° (c 1.0)
(MeOH), Anal. Calcd for C15H25NO4 × 0.1 H2O: C, 63.18; H,
8.91; N, 4.91. Found: C, 62.96; H, 9.01; N, 4.89; LCMS (m/z):
183 (M-Boc, 100), 305 (M + Na, 60).
(2R,3aR,6aR)-2-(tert-Butoxycarbonylamino)octahydro-
pentalene-3a-carboxylic Acid (4). A solution of 10 (166 g,
0.585 mol), MeOH (930 mL), and H2O (300 mL) was treated
with 10 M NaOH (234 mL) and the reaction was heated to 65
°C for 2 h. The reaction was cooled to 10 °C and acidified with
conc. HCl until pH = 3; H2O (1.8 L) was added to the slurry
and stirring continued for 1/2 h. The slurry was extracted with
CH2Cl2 (2 L; 2 × 1 L), washed with brine (1 L), dried over
Na2SO4, filtered and evaporated (29 in Hg, 40 °C) to afford 4
(146.1 g) as a white foam in 93% yield.
25
[α]D −39.64 (c 1.1, MeOH), Anal. Calcd for C15H24ClNO4:
C, 56.69; H, 7.61; Cl, 11.16; N, 4.41. Found: C, 56.80; H, 7.85;
Cl, 10.34; N, 4.43.
(1S,4S)-Methyl-4-(tert-butoxycarbonylamino)-1-(3-
chloropropyl)cyclopent-2-ene-carboxylate (18). 1H
NMR (400 MHz, DMSO-d6) 7.06 (d, J = 7.6 Hz, 1 H), 5.75
(s, 1 H), 5.67−5.74 (m, 2 H), 4.59 (d, J = 7.6 Hz, 1 H), 3.56−
3.66 (m, 5 H), 2.65 (dd, J = 8.1, 13.2 Hz, 1 H), 1.78−1.92 (m,
1 H), 1.56−1.77 (m, 3 H), 1.47 (dd, J = 7.0, 13.3 Hz, 1 H),
1.38 (s, 9 H); 13C NMR (100 MHz, DMSO-d6) 174.84, 155.00,
134.62, 134.33, 77.67, 58.29, 55.73, 54.87, 51.94, 51.85, 45.41,
1H NMR (400 MHz, DMSO-d6) 12.09 (s, 1 H), 6.80 (br. s.,
1 H), 3.75−3.92 (m, 1 H), 2.56−2.70 (m, 1 H), 2.11 (br. s., 1
H), 1.90 (br. s., 2 H), 1.72 (br. s., 1 H), 1.51−1.67 (m, 3 H),
1.37 (s, 11 H), 1.21 (br. s., 1 H); 13C NMR (101 MHz, DMSO-
d6) 179.05, 154.95, 77.38, 56.99, 50.59, 45.64, 42.63, 38.84,
38.51, 34.26, 28.18, 26.29; Anal. Calcd for C14H23NO4 × 0.07
CH2Cl2 × 0.14 H2O: C, 60.91; H, 8.49; N, 5.05; KF, 0.78.
25
45.24, 35.27. 28.23, 28.05; [α]D −112.61 (c 1.15, MeOH);
Anal. Calcd for C15H24ClNO4: C, 56.69; H, 7.61; Cl, 11.16; N,
4.41. Found: C, 56.59; H, 7.84; Cl, 11.08; N, 4.35.
(1R, 4S)-Methyl-4-(tert-butoxycarbonylamino)-1-(3-
iodopropyl)cyclopent-2-ene-carboxylate (13). A solution
of 17 (311 g, 0.978 mol) in acetone (3.7 L)was treated with
NaI (741 g, 4.89 mol, 5 equiv) portion-wise, and the reaction
was heated to 60 °C for 18 h (The progress of the reaction was
monitored by NMR (DMSO-d6 400 MHz) observing the
disappearance of the triplet at 3.58 ppm for 17 and the
appearance of the signal at 3.13 ppm for 13). The reaction was
cooled to rt, diluted with EtOAc (4 L), and washed with brine
(2 L). The organic was rotary evaporated (29 in Hg, 40 °C) to
a residue which contained water. EtOAc (1 L) was added to the
residue, and the layers were allowed to separate. The separated
organic layer was dried over Na2SO4, filtered, and concentrated
to afford 13 (386 g) in 96% yield. 1H NMR (400 MHz, CDCl3)
5.74−5.84 (m, 2 H), 4.69−4.98 (m, 2 H), 3.72 (s, 3 H), 3.14 (t,
J = 6.5 Hz, 2 H), 2.07−2.26 (m, 2 H), 1.83−1.98 (m, 1 H),
25
Found: C, 60.99; H, 8.54; N, 5.01; Karl Fischer, 0.79; [α]D
+13.94° (1.05) (MeOH); LCMS (m/z): 196 (M-Boc, 80), 291
(M + Na, 100).
tert-Butyl-(2R,3aR,6aR)-3a-(3-(Trifluoromethyl)-
5,6,7,8-tetrahyro-1,6-naphthyridine-6-carbonyl)-
octahydrentalen-2-yl-carbamate (21). A solution of 4 (326
g, 1.21 mol) and N,N-dimethylacetamide (3.3 L) was treated
with 1-hydroxybenzotriazole (217 g, 1.57 mol) and 5 (400 g,
1.45 mol). The slurry was cooled to 10 °C, followed by
portionwise addition of EDC (354 g, 1.82 mol) at <20 °C.
DIEA (634 mL, 3.63 mol) was added dropwise while
maintaining a temperature <20 °C. The reaction was stirred
for 2 h and quenched into ice-cold 5% NaHCO3 (8.7 L). The
above reaction mixture was stirred, and after a 15 min, a
precipitate formed, and the slurry was stirred for 1 h. The solid
was filtered and washed with aqueous 5% NaHCO3 solution (4
× 2 L), H2O (2 × 3 L), and heptane (2 × 2 L). The solid was
placed into vacuum oven (29 in Hg) for 18 h at 40 °C and
25
1.59−1.83 (m, 3 H), 1.44 (s, 9 H), [α]D −47.64° (c 1.5,
MeOH); Anal. Calcd for C15H24INO4 × 0.5 H2O × 0.28
EtOAc: C, 44.53; H, 6.11; I, 29.18; N, 3.22; KF, 0.21. Found:
C, 44.92; H, 6.39; I, 29.53; N, 3.36; KF, 0.20.
1
(2R,3aR,6aR)-Methyl-2-(tert-butoxycarbonylamino)-
octahydropentalene-3a-carboxylate (10). A solution of 13
(385 g, 0.941 mol) and toluene (8.5 L) was heated to 78 °C. In
a separate flask was added 2,2′-azo-bis-isobutyronitrile (16 g,
0.094 mol, 0.1 equiv), tris(trimethylsilyl)silane (377 mL, 1.213
afforded 21 (413.6 g) as an off-white solid in 75% yield. H
NMR (400 MHz, CDCl3) 8.70 (s, 1H), 7.69 (s, 1H), 4.90−
4.68 (m, 2H), 4.60 (d, J = 7.3 Hz, 1H), 4.13 (d, J = 6.6 Hz,
1H), 4.02−3.83 (m, 2H), 3.55 (t, J = 5.7 Hz, 1H), 3.12 (br. s.,
2H), 2.21 (dd, J = 5.7, 12.8 Hz, 1H), 1.95 (br. s., 1H), 1.92−
F
dx.doi.org/10.1021/op500265z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
1.78 (m, 3H), 1.78−1.49 (m, 5H), 1.38 (s, 9H); [α]D25 +47.69°
(c 0.98) (MeOH), Anal. Calcd for C23H30F3N3O3 × 0.35 H2O
× 0.25 DMAc: C, 59.67; H, 6.84; F, 11.66; N, 9.52; KF, 1.32.
Found: C, 60.05; H, 6.96; F, 11.27; N, 9.41; KF, 1.33; HPLC
(Zorbax SD): 5.97 min, 86%; LCMS (m/z): 353 (M-Boc, 100),
476 (M + Na, 20).
reaction was stirred for 0.75 h at 0 °C; HPLC showed 23 to still
be present (16.1%, compared to 79% for 2, along with 2.4%
trans isomer 24). A second charge of 3 (3.75 g) in 1,2-
dichloroethane (20 mL) was made. HPLC after 30 min showed
a reduction of 23 to 1.3%, and after allowing the reaction to
warm to rt overnight, the amount of 23 was reduced to 0.45%
(although the trans isomer 24 increased to 4.6%). The reaction
was chilled in an ice bath, aqueous NaOH (3N, 430 mL) was
added. The layers were separated, and the aqueous layer (pH
13) was extracted with DCE (1 × 100 mL), upon which a
second extraction with DCE (100 mL) produced an emulsion
that required filtration through Celite. The combined organic
layers were dried (Na2SO4) and filtered, and the volatiles were
evaporated under reduced pressure. The crude product was
obtained as a red thick oil (98.3 g, 111% isolated yield). The
crude products from two other similar runs were combined
(totaling 183 g) and sent for chromatographic separation: first,
using the reverse phase to remove a small amount of 23, and
the trans reductive amination isomer 24 (minor), followed by
Chiralpak AD, using 2:1:1 IPA/EtOH/heptane to remove the
cis-reductive amination product 25 and provided 150 g of red
oil (residual EtOH detected). After treating to high vacuum
with heat (5 mm, 50 °C) there was provided 2 (137.7 g) in
(2R,3aR,6aR)-2-Aminooctahydropentalen-3a-yl)(3-
(trifluoromethyl)-7,8-dihydro-1,6-naphthridine-6(5H)-
yl)methanone Dihydrochloride (22). CAUTION: non-
restricting venting should be used, as there was a vigourous
outflow of gas during this reaction. A mixture of 21 (222.7 g,
0.491 mol) in HCl in MeOH (1.25 M, 4 L, 4.91 mol) was
heated in stages, first to 40 °C in 10 min, then to 50 °C in 10
min, and finally to 60 °C in 10 min, at which point the reaction
was held for 40 min. Vigorous off-gassing ceased after 40 min.
The reaction was cooled and evaporated (10 mm Torr) to
dryness at 40 °C. MeOH (1 L) was added, the contents swirled
until dissolved, and the volatiles removed under vacuum. This
MeOH treatment was repeated one more time. The orange
contents were suspended in i-PrOH (620 mL) at 40 °C for 15
min, and heptane (1.75 L) was added to the suspension.
Cooling the bath with ice commenced, and the suspension was
swirled for 20 min at 21 °C. The light yellow solid was filtered,
washed with 20% IPA/heptane (500 mL), and allowed to air-
dry at rt overnight and afforded 22 (186.5 g) as a yellow solid
in 89% yield. mp 193.0−195.0 °C; 1H NMR (400 MHz,
DMSO-d6) 8.83 (s, 1H), 8.29 (br. s., 4H), 7.19 (br. s., 3H),
4.85 (d, J = 17.6 Hz, 1H), 4.74 (d, J = 18.1 Hz, 1H), 3.85 (br. s.,
2H), 3.56 (br. s., 1H), 3.46−3.30 (m, 1H), 3.05 (br. s., 2H),
2.40 (dd, J = 7.1, 13.0 Hz, 1H), 1.91−1.72 (m, 4H), 1.70−1.59
(m, 2H), 1.44−1.21 (m, 2H); 19F NMR (376 MHz, DMSO-
d6)−60.63; 13C NMR (101 MHz, DMSO-d6) 174.58, 143.16,
132.66, 130.30, 124.94, 123.45, 123.13, 122.24, 96.23, 89.20,
58.01, 50.62, 45.46, 41.92, 38.67, 35.78, 32.46, 31.19, 26.45;
Anal. Calcd for C18H22F3N3O × 2HCl × 0.35 H2O: C, 49.97;
H, 5.75; F, 13.17; Cl, 16.39; N, 9.71; 1.46; Found: C, 50.04; H,
5.73; F, 12.74; Cl, 16.61; N, 9.67; KF, 1.36; HPLC (eclipLO):
8.76 min, 98%; LCMS (m/z): 354 (M + H, 100).
((2R,3aR,6aR)-2-((3R,4R)-3-Methoxytetrahydro-2H-
pyran-4-ylamino)octahydropentalen-3a-yl)(3-(trifluoro-
methyl)-7,8-dihydro-1,6-naphthyridin-6(5H)-yl)-
methanone (2). A solution of 22 (80 g, 0.187 mol), 1,2-
dichloroethane (400 mL), and water (140 mL) at 0 °C was
neutralized by careful addition of aqueous NaOH (1M, 140
mL) over 10 min, at <6 °C. The mixture was stirred for 2 min,
and an attempt was made to separate the layers; water (50 mL)
and brine (50 mL) were added, but a clean interface could not
be achieved. The biphasic mixture was filtered with Celite. A
minimum of dichloroethane should be used here to rinse the
filter-aid. The aqueous layer was drawn off and shaken with 1,2-
dichoroethane (200 mL), and this mixture was filtered through
the original Celite pad. The combined organic layers were
washed with aqueous NaHCO3 (saturated, 25 mL), dried
(Na2SO4), and filtered (washing the drying agent with 2 × 25
mL fresh DCE), and the solution containing 23 was used
directly in the next reaction. A KF analysis of the solution was
found to be 1.99% water.
1
75% yield. H NMR (400 MHz, CDCl3) 8.70 (s, 1H), 7.68 (s,
1H), 4.78 (d, J = 6.4 Hz, 2H), 4.11−3.81 (m, 5H), 3.74 (s,
2H), 3.61−3.47 (m, 2H), 3.43−3.23 (m, 8H), 3.12 (br. s., 2H),
2.82−2.71 (m, 1H), 2.21−1.85 (m, 3H), 1.82−1.48 (m, 15H),
1.35−1.20 (m, 1H). HPLC (EclipLO): 9.40 min, 98.2%.
((2R,3aR,6aR)-2-((3R,4R)-3-Methoxytetrahydro-2H-
pyran-4-ylamino)octahydropentalen-3a-yl)(3-(trifluoro-
methyl)-7,8-dihydro-1,6-naphthyridin-6(5H)-yl)-
methanone Semisuccinate (1). A solution of 2 (137.7 g,
0.284 mol) and acetone (660 mL) was warmed to 50 °C, and
succinic acid (34.9 g, 0.295 mol) was added in one portion. The
temperature jumped to 55 °C, the solution cleared immedi-
ately, and after settling back to 50 °C, no precipitate was
evident. The solution was seeded, and within a few minutes,
cloudiness ensued, followed by precipitation. The heating was
continued for 1.25 h total. The heat was removed, and the
suspension was allowed to come to rt over ∼1 h. The
suspension was chilled in an ice bath to 0 °C and filtered
through paper, and the solid was washed with ice-cold acetone
(150 mL), followed by room temperature acetone (50 mL).
After drying the isolated solid in a vacuum oven (29 mmHg, 27
°C), the batch was deemed unacceptable for delivery because of
the ivory-tinge color (139.2 g) and was not identical to the
brilliant white color of lots delivered previously.
Decolorization Procedure. A solution of 1 (139.2 g) and
MeOH (2.1 L) was treated with Darco G-60 charcoal (Aldrich
cat. # 242276−250 g, Lot # 04916KD) and the contents
warmed with 50 °C for 10 min. Hot filtration through Celite
proceeded, followed by washing the plug with MeOH (rt, 200
mL). The filtrate was evaporated (29 in Hg, 40 °C), and the
resulting white foam (159 g) was suspended in acetone (720
mL) and swirled at 50 °C for 45 min. In about 10 min of
swirling the hazy liquid, crystals started to form on the top of
the liquid surface; in 20 min, copious solid formed, and within
30 min, a solid was thick and required mechanical scraping to
free from the sides of the flask. After 45 min, the heat was
turned off; ice was added to the bath, and (when the flask was
cold to the touch) the solid was filtered, washed with rt acetone
(150 mL), and the solid dried in a vacuum oven (27 °C, 30″
Hg) for 60 h. The white solid 1 (135.1 g) was obtained in 81%
A solution of 23 (675 mL solution, assumed 66.5 g, 0.187
mol) was chilled to 0 °C. Acetic acid (glacial, 22 mL) was
added, followed by solid Na(AcO)3BH (56.05 g, 0.264 mol),
and the resulting suspension was stirred for 5 min. A solution of
3 (33.05 g, 0.254 mol, 86.5% ee) in 1,2-dichloroethane (200
mL) was added over 20 min, and a rise of 4 °C was noted. The
G
dx.doi.org/10.1021/op500265z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
1
S. J. C.; Chaplin, D. A.; Casy, G.; McCague, R. Tetrahedron Asymm.
2001, 703.
yield. mp: 166.5−168.6 °C; H NMR (400 MHz, DMSO-d6)
10.81−9.06 (m, 2H), 8.78 (s, 1H), 8.19 (br. s., 1H), 4.90−4.61
(m, 2H), 3.95 (d, J = 12.5 Hz, 1H), 3.85 (br. s., 2H), 3.80−3.60
(m, 1H), 3.54−3.32 (m, 3H), 3.32−3.15 (m, 5H), 3.01 (br. s.,
3H), 2.42−2.23 (m, 5H), 2.04−1.81 (m, 2H), 1.78−1.68 (m,
2H), 1.67−1.48 (m, 4H), 1.45−1.33 (m, 1H), 1.33−1.14 (m,
1H); 19F NMR (376 MHz, DMSO-d6) −60.6 ppm; 13C NMR
(101 MHz, DMSO-d6) 175.22, 174.26, 143.91, 131.69, 129.88,
125.16, 123.14, 122.82, 122.50, 122.46, 74.14, 64.89, 64.83,
57.82, 55.61, 55.09, 53.54, 45.34, 43.91, 39.00, 37.21, 32.76,
30.34, 27.13, 26.50; C28H38F3N3O7: C, 57.43; H, 6.54; F, 9.73;
N, 7.18; Found: °C, 57.33; H, 6.53; F, 9.50; N, 7.09; KF <0.1%;
(9) Zheng, C.; Cao, G.; Xia, M.; Feng, H.; Glenn, J.; Anand, R.;
Zhang, K.; Huang, T.; Wang, A.; Kong, L.; Li, M.; Galya, L.; Hughes,
R. O.; Devraj, R.; Morton, P. A.; Rogier, D. J.; Covington, M.;
Baribaud, F.; Shin, N.; Scherle, P.; Diamond, S.; Yeleswaram, S.; Vaddi,
K.; Newton, R.; Hollis, G.; Friedman, S.; Metcalf, B.; Xue, C.-B. Bioorg.
Med. Chem. Lett. 2011, 1442.
(10) (a) Jiao, R.; Morriello, G.; Yang, L.; Goble, S. D.; Mills, S. D.,
Pasternak, A.; Zhou, C.; Butora, G.; Kothandaraman, S.; Guiadeen, D.;
Moyes, C.; Tang, C. US Patent No. 6,812,234 B2, November 2, 2004.
(b) The deutero-enriched form was disclosed in Czarnik, A. W. US
Patent Application No. US 2009/0076065, March 19, 2009.
(11) We were unaware of this incompatibility as it appeared in
Greene and Wutz, Protective Groups in Organic Synthesis, 3rd ed.;
Wiley-Interscience: New York, pp 494−653.
(12) We would like to thank Dr. Wim Aelterman for these
suggestions.
(13) Commercially available from Anichem, North Brunswick, NJ,
Cat. No. H10367.
(14) Mills, J. E.; Maryanoff, C. A.; McComsey, D. F.; Stanzione, R.
C.; Scott, L. J. Org. Chem. 1987, 52, 1857.
25
[α]D +26.5° (c 1.0) (MeOH); HPLC (eclipLO): 9.33 min,
99.1%; HPLC (genpur01): 13.17 min, 96.1%; Chiral HPLC
(xlb5): 6.31 min, 99.5%.
ASSOCIATED CONTENT
■
S
* Supporting Information
Copies of 1H and 13C NMR for all reported compounds and all
HPLC and chiral HPLC methods. This material is available free
of charge via the Internet at http://pubs.acs.org.
(15) Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C.
A.; Shah, R. D. J. Org. Chem. 1996, 61, 3849.
AUTHOR INFORMATION
Corresponding Author
*E-mail: cteleha@its.jnj.com. Fax: 215-540-4611. Tel.: 215-
628-5225.
■
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) Cai, C.; Kang, F.-A.; Hou, C.; O’Neill, J.; Opas, E.; McKenney, S.;
Johnson, D.; Sui, Z. Bioorg. Med. Chem. Lett. 2013, 23, 351.
(2) See companion paper describing the preparation of an bicyclic
ether in: Teleha, C. A.; Branum, S.; Zhang, Y.; Reuman, M. E.; Van der
Steen, L.; Verbeek, M.; Fawzy, N.; Leo, G. C.; Winters, M. P.; Kang,
F.-A.; Kolpak, M.; Beauchamp; D. A.; Lanter, J. C.; Russell, R. K.; Sui,
Z.; Vanbaelen, H. Org. Process Res. Dev. 2014, 10.1021/op500266w.
(3) (a) Jensen, M.; Larsen, R.; Sidler, D. R. WO 2005044264, 2005.
(b) Butora, G.; Guiadeen, D.; Kothandaraman, S.; Maccoss, M.; Mills,
S. G.; Yang, L. WO 2005072361, 2005. (c) Cai, D.; Fleitz, F.; Ge, M.;
Hoerrner, S.; Javadi, G.; Jensen, M.; Larsen, R.; Li, W.; Nelson, D.;
Szumigala, E.; Yang, L.; Zhou, C. WO2005044795, 2005. (d) Kuduk,
S. D.; Di Marco, C. N.; Greshock, T. J. WO2012158473, 2012.
(d) Kosjek, B.; Nti-Gyabaah, J.; Telari, K.; Dunne, L.; Moore, J. C.
Org. Process Res. Dev. 2008, 12, 584. For a chiral method to the OBz
derivative, see: Jadhav, M. S.; Righi, P.; Marcantoni, E.; Bencivenni, G.
J. Org. Chem. 2012, 77, 2667.
(4) For a biocatalytic method to a useful intemediate with the desired
chirality, see: Kuduk, S. D.; Skudlarek, J. W. WO2011159553, 2011.
For reagent-based method using a catalytic oxazaborolidine borane
reduction to introduce the required chirality, see: Reuman, M.; Zhang,
Y.; Kang, F.-A.; Branum, S.; Teleha, C. A., manuscript in preparation.
(5) Cai, C.; Kang, F.-A.; Beauchamp, D. A.; Sui, Z.; Russell, R. K.;
Teleha, C. A. Tetrahedron: Asymm. 2013, 24, 651.
(6) Beckwith, A. L. J.; Roberts, D. H. J. Am. Chem. Soc. 1986, 108,
5893.
(7) For other examples of radical cyclizations: for [5.3.0] systems see
(a) Kim, S.; Kim, K. H.; Cho, J. R. Tetrahedron Lett. 1997, 3915. for
general review see Org. Reactions, vol. 48, 1996; for formation of
bicyclic alkanes by polycyclizations see (b) Kim, S.; Cho, C. H. Can. J.
Chem. 2005, 917. (c) Kim, S.; Cho, C. H.; Lim, C. J. J. Am. Chem. Soc.
2003, 9574. For [4.3.0] systems see (d) Clive, D. L. J.; Sunasee, R.;
Chen, Z. Org. Biomol. Chem. 2008, 2434−2441. (e) Clive, D. L. J.;
Sunasee, R. Org. Lett. 2007, 2677.
(8) Qiu, J.; Pingsterhaus, J. M.; Silverman, R. B. J. Med. Chem. 1999,
4725. Smith, M. E. B.; Lloyd, M. C.; Derrien, N.; Lloyd, R. C.; Taylor,
H
dx.doi.org/10.1021/op500265z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX